Llumnix: Dynamic Scheduling for Large Language Model Serving.
Biao SunZiming HuangHanyu ZhaoWencong XiaoXinyi ZhangYong LiWei LinPublished in: OSDI (2024)
Keyphrases
- language model
- dynamic scheduling
- language modeling
- flexible manufacturing systems
- probabilistic model
- n gram
- speech recognition
- document retrieval
- retrieval model
- query expansion
- context sensitive
- information retrieval
- test collection
- language modelling
- mixture model
- statistical language models
- process planning
- language model for information retrieval
- manufacturing systems
- document length
- smoothing methods
- ad hoc information retrieval
- grid services
- translation model
- query processing
- relevance model
- query terms
- dirichlet prior
- image analysis